The Willamette River Case: Costs 207 



differences in economic efficiency were this method of undertaking 

 river basin projects employed, there are substantial reasons for 

 believing that there would be differences in the distribution of 

 income resulting from the project. 



Differences in Costs Under Public and Private Development 



If the reimbursable features of a multiple purpose project are 

 financed by nonfederal participants in the development, two major 

 differences in income redistributive effects become probable. First, 

 the accounting costs will differ as among the alternative ways of 

 financing the reimbursable feature; moreover, these costs will 

 exhibit differences in their incidence. Second, there are possibilities 

 of differences in the distribution of gains as well as costs, if the 

 reimbursable feature is financed by other than a public nonfederal 

 participant; this is caused, at least in part, by the suspension of the 

 preference provision in the distribution of the project's electrical 

 output. 



In looking at the differences in accounting costs, we assume 

 alternatively that the reimbursable feature is financed by the federal 

 government, a nonfederal public body, and a private utility. Ac- 

 counting costs associated with each alternative are shown in Table 

 38.^ Here we assume that the separable and common costs of the 

 hydroelectric features of a multiple purpose project would approxi- 

 mate $280 per kilowatt of capacity. We assume a relatively small- 

 scale development — 80,000 kilowatts — which is not unrealistic for 

 a Willamette River Basin hydroelectric site. Construction costs are 

 assumed to be the same, irrespective of the means by which the 

 investment funds for the power feature are provided. Total in- 

 vestment differs among the three alternatives, however, because of 

 assumed differences in relevant rates of interest over the construc- 

 tion period. The differences in investment outlays, however, are 

 not very significant for a project which would require a relatively 

 short time to build. 



' These estimates are only approximations of the magnitudes which would 

 appear for corresponding items in any particular case, as there will be consider- 

 able variability both in accounting practice and financial characteristics among 

 particular installations. The estimates, however, represent reasonable approxi- 

 mations based on either average data or recommended practices as referenced 

 in notes to Table 38. 



